Quoting Bernhard Eversberg <ev_at_BIBLIO.TU-BS.DE>:
>
> The complete table with evaluation and comments:
> http://www.allegro-c.de/google/gbscheck.htm
This study is highly valuable, Bernhard. What I think many do not
understand is the length of the 'long tail' for books and other
publications. Yet, once Google has scanned some percentage of books,
that's all that most users will ever access. For those who tout that
Google will provide greater access to the public, the truth is that
they will provide greater access to some percentage of the world's
books, and no access to the rest. Not that this is much worse than the
fact that many books in libraries are hidden away as well.
kc
>
> Just the figures:
>
> 56 Titles with one complete digitized copy
> 8 with more than one
> 16 Full text of other edition
> 6 Part of full text available (only newer titles)
> ---------
> 86 Titles with at least some full text (most of them 19th cent.)
>
> 23 Snippets (i.e., a digitization *exists*)
> (These may presumably become accessible eventually, after an
> agreement with publishers)
>
> 354 Metadata only
>
> 37 Titles not found in GBS
> ---------
> 500
>
> We can, however, *NOT* conclude that 17.2% of total borrowings could be
> satisfied by using GBS since, other than in the sample, the majority of
> books borrowed is 20th cent. and later (where GBS coverage is
> considerably smaller):
> In the total number of recorded borrowings, we found
> 1 % before 1900
> 80 % 1900-1999
> 19 % 2000-
>
> Only 24 full text instances are after 1899, about 8% of borrowings
> within that period. Currently mostly just snippets, of course!
> To be more realistic, the sample should thus have covered time periods
> in a very different way...
>
>
> B.Eversberg
--
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet
Received on Mon Mar 29 2010 - 12:10:18 EDT